Site News  |   Monitor Keywords  |   Monitor Archive  |   Organizer  |   Account Info  |

Method for rapid identification of alternative splicing

The present invention relates to a method for isolating and identifying alternatively spliced mRNA.

The number of proteins produced by the human genome likely numbers in the hundreds of thousands. However, recent evidence indicates that the human genome contains only 30,000 to 45,000 different genes. Clearly, each gene is producing multiple proteins.

Alternative splicing of primary RNA transcripts is a major mechanism for increasing production of proteins from the human genome. It is known that 30% to 60% of genes undergo alternative splicing to produce messenger RNA (mRNA). Modrek B et al. Nat. Genet. 30, 13-19 (2002). These alternatively spliced mRNA are translated into alternative splice form proteins that contain amino acid sequences different than the corresponding protein produced by normally spliced mRNA.

Alternative splice form proteins are often expressed in a tissue-specific manner, or under certain physiologic or disease states. Modrek B et al., Nucl. Acids Res. 29, 2850-2859 (2001). Consequently, certain alternatively spliced mRNA are present in a limited number of cells in a subject suffering from a given disease or condition. For example, it is known that many types of cancer cells produce alternative splice forms which are not found in normal cells from the same subject. Cancer-associated genes such as CD44 (Rodriguez C et al., Int. J. Cancer 64, 347-354, 1995), estrogen receptor (Castles C G et al., Cancer Res. 53, 5934-5939, 1993), FGF receptor (Luqmani Y A et al., Int. J. Cancer 64, 274-279, 1995), DNA polymerase (Bhattacharyya N et al., DNA Cell Biol. 18, 549-554, 1999), cathepsin B (Gong Q et al., DNA Cell Biol. 12, 299-309, 1993), FHIT (Panagopoulos I. et al., Cancer Res. 56, 4871-4875, 1996), BRCA1 (Thakur S et al., Mol. Cell Biol. 17, 444-452, 1997) and BRCA2 (Bieche I et al., Cancer Res. 59, 2546-2550, 1999), produce alternatively spliced mRNA that are specifically expressed in cancerous tissues. Other disease states in which alternative splice forms are specifically produced in certain tissues include diabetes, Alzhiemer's disease and systemic lupus erythematosus (SLE).

Drugs that target proteins specific to cancerous or other disease tissue have proven efficacious in the appropriate patient population. For example, successful treatment of breast cancer has been reported for drugs which target the estrogen receptor (Jordan C, Clin. Ther. 24 Suppl A, A3-16, 2002) or the HER-2 receptor (Thomssen C, Anticancer Drugs 12 Suppl 4, S19-S25, 2001; Yip Y L et al., Cancer Immunol. Immunother. 50; 569-587, 2002). The genetic alterations present in tumor-specific proteins, such as mutations in p53, BRCA 1 and BRCA2, provide another source of targets. Thus, the proteins produced from alternatively spliced mRNA produced specifically in cancers or other disease states are also attractive therapeutic targets.

However, proteins produced from alternatively spliced mRNA have not been widely exploited as therapeutic targets. The major impediment to using such proteins as therapeutic targets has been the incidental or tedious nature by which alternatively spliced mRNA are found. Present methodologies are limited to either cDNA cloning (which is highly labor intensive) or RT/PCR (which focuses only on known portions of genes). In addition, most cloning- and RT/PCR-based methods are highly biased, as they require prior knowledge of the alternatively spliced mRNA sequence.

An unbiased procedure for discovery of alternatively spliced mRNA has been reported in U.S. Pat. No. 6,251,590 of Schweighoffer et al. However, the Schweighoffer et al. method identifies only the region in the alternatively spliced mRNA that is different from the normally spliced mRNA. The cDNA corresponding to both the normal and alternatively spliced mRNA must be separately cloned in order to pinpoint the alternatively spliced region in the context of the full-length molecule. The sequencing of multiple cDNA clones is also required to determine the prevalence of a given alternatively spliced mRNA. The Schweighoffer et al. method thus required a substantial investment of both time and resources in order to identify alternatively spliced molecules.

Thus, an unbiased method of rapidly and easily identifying alternatively spliced RNA in biological sample is needed, in which both the full-length normal and alternatively spliced mRNA are simultaneously isolated for comparison. Ideally, such a method would not rely on multiple cloning and sequencing steps for determining the identity and relative abundance of alternative splice forms in a given sample.

The present invention is directed to an unbiased method for isolating and identifying full-length alternatively spliced RNA, wherein the alternatively spliced RNA is isolated in conjunction with its counterpart normally spliced RNA. The practice of this method thus does not require foreknowledge of either the normal or alternatively spliced RNA sequences, or the nature of the alternative splice. The method also does not require multiple cloning or sequencing steps in order to identify the alternatively spliced RNA.

The invention provides a method of identifying an alternatively spliced RNA by comparing populations of cDNA molecules obtained from two biological samples. One sample represents a first physiological condition, and the other sample represents a second physiological condition. The two cDNA populations are separately tagged with different compounds, and denatured portions of each tagged cDNA population are annealed to each other under conditions which allow the formation of a mixed population of cDNA molecules. This mixed population comprises single-stranded cDNA molecules from both populations, double-stranded cDNA comprising cDNA molecules from only the first or second cDNA populations, and double-stranded cDNA comprising cDNA molecules from both the first and second cDNA populations.

Double-stranded cDNA comprising cDNA molecules from both the first and second cDNA populations are isolated from the mixed population by first selecting for those molecules comprising the tag specific to the first cDNA population, followed by selecting for molecules which also contain the tag specific to the second cDNA population. Alternatively, double-stranded cDNA comprising cDNA molecules from both the first and second cDNA populations can be isolated by selecting for molecules comprising the tag specific to the second cDNA population, followed by selecting for molecules comprising the tag specific to the first cDNA population.

The double-stranded cDNA selected above comprises two types. The first type comprises two cDNA molecules with perfectly matched sequences, in which each cDNA molecule represents normally spliced mRNA. The second type comprises two cDNA molecules with at least one area of mismatched sequence. In the second type of double-stranded cDNA, one cDNA strand represents the alternatively spliced mRNA molecule and the other cDNA strand represents the normally spliced counterpart of the alternatively spliced mRNA.

The mismatched sequence is unpaired with respect to the opposite strand and comprises a single-stranded region in the otherwise paired sequences. Such a double-stranded cDNA encompassing a mismatched sequence is then isolated with reagents which bind to regions of single-stranded nucleic acid. The two nucleic acid strands of said selected double-stranded cDNA are coupled, yielding a single molecule that can be analyzed to identify the normal and alternatively spliced molecules.

A kit comprising some or all of the components and for performing the present method, along with instructions for their use, is also provided.

FIG. 1 is a flow chart of a method according to the invention.

FIGS. 2A-2E are diagrams showing the isolation and identification of alternatively spliced RNA according to one embodiment of the invention.